[Bug nptl/13165] pthread_cond_wait() can consume a signal that was sent before it started waiting

[bugdal at aerifal dot cx]

http://sourceware.org/bugzilla/show_bug.cgi?id=13165

--- Comment #29 from Rich Felker <bugdal at aerifal dot cx> 2012-09-20 20:31:26 UTC ---
The lack of language to explicitly prevent something is not necessary. Do you
see any language that explicitly prevents an implementation from writing the
text "42" to stdout when you call strlen? I would grant that Torvald has an
argument if the application had called any glibc functions not specified by
POSIX (which could be defined by the implementation to do all sorts of crazy
things) or if the application had passed constants other than those defined by
POSIX to standard functions (e.g. a special attribute for the condition
variable). But in the absence of that, no interface in the standard library can
have further side effects on other interfaces/objects than what it's specified
to do.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.